SARS-CoV-2 Nsp14-MTase inhibitors

Abstract

The emergence of SARS-CoV-2, the causative agent of COVID-19, has highlighted theneed for advanced antiviral strategies. Targeting the coronaviral methyltransferasensp14, which is essential for RNA capping, offers a promising approach for thedevelopment of small-molecule inhibitors. We designed and synthesized a series ofadenosine 5′-carboxamide derivatives as potential nsp14 inhibitors and identifiedcoumarin analogs to be particularly effective. Structural modifications revealed theimportance of the 5′-carboxyl moiety for the inhibitory activity, showing superiorefficacy compared to other modifications. Notably, compound 18l (HK370) demonstratedhigh selectivity and favorable in vitro pharmacokinetic properties and exhibitedmoderate antiviral activity in cell-based assays. These findings provide a robustfoundation for developing targeted nsp14 inhibitors as a potential treatment forCOVID-19 and related diseases.

Introduction

Viruses belonging to the family Coronaviridae (order Nidovirales) pose a significantpandemic threat, as was demonstrated by SARS-CoV, MERS-CoV and, most notably, bySARS-CoV-2 in 2002, 2012 and 2019, respectively.1 The genetic information of SARSCoV-2 consists of a large (∼30 kb) positive-sense, single-stranded RNA (+ssRNA), whichencodes 4 structural proteins, 16 non-structural proteins and several accessoryfactors.2 The virus replicates in double-membrane vesicles (DMVs) derived from thehost endoplasmic reticulum and, therefore, does not have access to the host's mRNAcapping machinery.3 The 5′ end of the eukaryotic mRNA is equipped with a cap-1 orcap-2 structure, which is important for several cellular processes including translationand self-recognition. The absence of this cap structure triggers the activation ofcytosolic sensors (e.g., IFIT1 or MDA5) and subsequently initiates an immune response.4,5To mimic the host's mRNA cap, SARS-CoV-2 utilizes its own capping enzymes, includingtwo MTases – nsp14 (N7 methylation; cap-0) and nsp16 (2′O methylation; cap-1). BothMTases are SAM-dependent, which makes them a suitable target for small-moleculeinhibitors.

Coronaviral MTases are a focal point of medicinal chemistry research since the beginningof the SARS-CoV-2 pandemic.7–21 Otava et al. described SAH analogs with a modifiednucleobase (1) targeting the lateral cavity above the SAM-binding site7 and variousreplacements of the amino acid moiety on the 5′ end were explored by severalgroups.11,22–26 A significant portion of this research has focused on arylsulfonamidesderived from 5′-aminoadenosine (2, 3), with the sulfonamide moiety being crucial for theinhibitory activity due to its specific geometry.24–28 Derivatives of adenosine-5′-carboxylic acid represent an under-explored area in the search for coronaviral nsp14inhibitors. To date, only one such compound (4) has been described in the literature11and is considered inferior due to its poor inhibitory activity against SARS-CoV-2 nsp14(IC50 = 12 μM).

In this work, we synthesized a series of amides derived from adenosine-5′-carboxylicacid leading to the development of nanomolar inhibitors with general structures 5 and 6.We further examined the importance of the amidic moiety for inhibitory activity andtested our compounds in a cell-based assay (Fig. 1).

‍

Results and discussion

Synthesis

We started the synthesis from adenosine (7), which was isopropylidene-protected toafford 8.29 Subsequent oxidation of the 5′ carbon using TEMPO/PhI(OAc)2 yieldedadenosine 5′-carboxylic acid 9.30 Treatment of 9 with SOCl2 afforded highly reactiveintermediate 10, which was used without any purification in amidic coupling with aselected amine leading to amides 11. Final removal of the isopropylidene protecting groupwas achieved using 80% formic acid yielding 12.

Synthesis of analogs with a modified nucleobase started from 2′,3′-protected 7-iodotubercidine 14, which was prepared according to a published procedure.31 Oxidationof the 5′ carbon was again achieved with TEMPO/PhI(OAc)2 yielding 15.30 A differentapproach was used for synthesis of 16, as some amines reacted poorly with acyl chloride10. Propanephosphonic acid anhydride (T3P) mediated peptide coupling between theamine of choice and the acid 15 smoothly afforded products in 2 to 24 hours inmoderate to good yields. Installation of the 5-ethynylpyrimidine moiety in position 7 ofthe nucleobase was achieved via the Sonogashira cross-coupling,7 leading to 17 whichwas subsequently treated with 80% formic acid to afford final compounds 18 (Scheme 1).

To explore other linkers, we mesylated compound 8 and, via nucleophilic substitutionwith NaN3, prepared compound 19. Subsequent hydrogenation yielded compound 20,which was then used for the preparation of amides (21a, 21k) and sulfonamide(22k).32 Finally, commercially available compound 23 was treated with 7-mercapto-4-methylcoumarin to afford compound 24. Oxidation of the sulfur linker using Oxone®yielded sulfone 25 (Scheme 2).33

Conslusion

In this work, we describe the discovery of SARS-CoV-2 nsp14 inhibitors with novelstructural motifs based on adenosine 5′-carboxamides derived from bicyclic amines.Through scaffold hopping, we found out that coumarin analogs are particularly effective.We also explored the importance of the 5′-carboxyl moiety for inhibitory activity.Analogs bearing a methylated amide, reversed amidic linker, sulfonamide, sulfide, orsulfone showed a significant decrease or complete loss of the inhibitory activity.Compound 18l (HK370), with an IC50 value of 31 nM, exhibited high selectivity for nsp14over human RNMT (IC50 (RNMT) > 25 μM; CC50 > 50 μM), a favorable in vitro metabolicprofile and good transepithelial permeability. Notably, our inhibitors show moderateefficacy in a cell-based assay (EC50: 12 ± 6 μM, Calu-3 cell line). Overall, this workprovides a strong foundation for the development of targeted nsp14 inhibitors aspotential treatments for COVID-19 and other coronavirus-related diseases.

‍

Read the full paper here